A non-maximum suppression process is often used with edge detection techniques and with computer vision tasks for object detection and tracking. The non-maximum suppression process sets all current pixel values to zero in a neighborhood around each current pixel if the current pixel is not a local maxima value. If the local maxima value is an absolute maximum value in the neighborhood, the local maxima value is a true result of the non-maximum suppression process.
It would be desirable to implement a non-maximum suppression in image data.